It's Complicated: Why Aducanumab's Approval Raises More Questions Than Answers

Alzheimer disease experts Richard S. Isaacson, MD, and Marc E. Agronin, MD, discuss the recent aducanumab approval and some of the clinical community's concerns.

The FDA's decision to approve Biogen's anti-amyloid therapy aducanumab (Aduhelm) for the treatment of Alzheimer disease is still sinking in for many--clinicians and patients alike. What has seemed at times to be only a far-off fantasy is now in our hands; a treatment with the potential to alter the course of the devastating neurodegenerative disease.

It's no surprise, then, that this moment that so many have longed for has generated a spectrum of reactions across the Alzheimer disease community. Ripe for critique, the ground-breaking approval comes after a tumultuous clinical trial and regulatory journey filled with stops and restarts, many of which have raised concerns over data validity.

Beyond the challenges posed by the EMERGE and ENGAGE data analyses, the first-of-its-kind approval, based on a "surrogate end point that is reasonably likely to predict a clinical benefit," has raised many other questions about patient selection, access to the therapy, and diagnosis of the disease itself.

To better understand the current feelings of the clinical community and the opportunities—and challenges—that this approval promotes, NeurologyLive brought together 2 clinician researchers with expertise in the diagnosis and treatment of Alzheimer disease: Richard S. Isaacson, MD, director, Alzheimer's Prevention Clinic at Weill Cornell Medicine and NewYork-Presbyterian and assistant dean of Faculty Development at Weill Cornell in New York, NY; and Marc E. Agronin, MD, senior vice president of behavioral health and chief medical officer of the MIND Institute at Miami Jewish Health in Miami, Fl. Watch them discuss the implications of aducanumab's approval below or read the provided video transcript.


Richard S. Isaacson, MD: Hello and thank you for joining this NeurologyLive Peers & Perspectives presentation entitled "Advances in the Treatment of Alzheimer Disease." Today we're going to be discussing the recent FDA approval of aducanumab and the current status of drugs in development for the treatment of Alzheimer. My name is Dr. Richard Isaacson, I'm director of the Alzheimer's Prevention Clinic at Weill Cornell Medicine in New York Presbyterian in New York City. I'm also an assistant dean of Faculty Development at Weill Cornell, and today, I'm very privileged and looking forward to this chat. I'm here with Dr. Marc Agronin. He's the senior vice president of behavioral health and Chief Medical Officer of the MINDS Institute at Miami Jewish Health in Miami, Florida. Thanks so much for joining. Let's begin. Dr. Agronin, we've been on this road together for over 15 years or so. When I was back in Miami, you were the one I referred my patients to for all the clinical trials, for the amyloid studies. We've had patients in these amyloid studies for 15 years. What a week it's been with the approval of aducanumab. I don't even know where to begin, because it's such a such a milestone in our field, but when you heard aducanumab was approved, what was the very first thing to go through your mind?

Marc E. Agronin, MD: Well, first of all, Richard, thank you for the invitation to be here. And you're right; we've worked together a long time with many, many mutual patients. And so it's really been a long journey. I think my initial reaction this week was one of hope that we are making some moves in terms of being able to offer disease modification to patients even though we know that there's a lot of questions, some controversy over this. I think as you mentioned before, we've been on a long journey here, reflecting back 15 years ago, we knew back then, you know the scope of Alzheimer disease, but we really struggled to accurately diagnose it. We knew that immunotherapy played a role with these transgenic mice that we could see it cleared amyloid from their brains, they seem to get better, and we were on on the verge of these new immunotherapy trials in humans, there was a lot of excitement. Now, 15 years later, we have something on the market and I think we're excited. We're hopeful. But there's a lot to talk about, and a lot of issues we still need to address here.

Richard S. Isaacson, MD: Yeah, I think the same exact thing from my end. I didn't know, I couldn't make predictions. I just didn't know what was going to happen. I think most people didn't know; I think some people felt one way other people felt another way. When it comes to the quality of scientific data, in having done many investigator-initiated trials where I've created the trial, I've written the IRB, I've written the study protocol, I've, you know, gone for months or years, sometimes with statisticians, I understand how hard it is to get quality research done. I have empathy. Even when you have billions of dollars or millions of dollars, whatever it is to invest in a clinical research pipeline and clinical trials, it is hard to get studies done. And you know, how valid is the data? What's the data quality? We can argue about that a lot and we can talk about the studies and what one study showed and what the other did, and what the statistical analyses showed--we can talk about that, but I feel like we're on a different side now. We now have a drug that's approved for patients. So I feel like we should probably focus more on what does this mean? The drug was approved in the FDA labeling and the package insert for the treatment of Alzheimer disease. Full stop. The drug was approved based on reducing amyloid in the brain. Interesting.

Marc E. Agronin, MD: The FDA has done something a little bit different this time because especially when an initial label has started to come out, we all realize that in part, this is really focused on amyloid reduction. And to some extent, over time, it's going to stand or fall on the integrity of the amyloid hypothesis for Alzheimer disease. We know from the very first study that Alois Alzheimer did on his patient where he looked at brain tissue, we know that amyloid and tau, these plaques and tangles are the hallmarks of this disease. The question all long has been, are they the only hallmarks? And if you get rid of one or both of them, what does that do for the disease? We've seen clearly through research that these medications can reduce amyloid burden, we know that. The real question is what does that mean? I sometimes use the analogy of, you know, if your house is on fire, and you put out the fire, maybe that's going to stop any further problems, but you've got a lot of damage there, and we know that's the case with brains with Alzheimer disease. And for that matter, for other forms of neurocognitive disorders. By the time they walk into our office, there's so much damage in the brain, the question is, what can we possibly do to mitigate that? So it's a dilemma, and we're still going to see over time with the ongoing EMBARK trial, which is still looking at individuals on aducanumab gathering data, that's going to really help tell the full story over the coming years.

Richard S. Isaacson, MD: This is the first drug approved in almost 20 years--I think it's 18 years or so. When you look at the failures, you look at the minimal success. This is the first time I've actually gone back and read a package insert in a while. I remember package inserts being quite long. I remember they would unroll them and you know, read every word in small print. This one I had a PDF, so I could make it larger. And you know, I stopped several times during reading it because I didn't know what the package insert would say? Would there be a requirement for an amyloid test to prove that the person has amyloid if you're going to be given an amyloid drug? That's one thing. Well, there wasn't that written in the package insert. So now is that the responsibility of the physician? That's one thing I wanted to chat with you about. Number 2, it's for the treatment of Alzheimer disease. You know, when I think about the term Alzheimer disease, I think of stage 1, 2, and 3. Stage 1, pre-clinical, pre-symptomatic Alzheimer where there's amyloid in the brain but no symptoms yet. Forty-six million Americans are estimated to have pre-clinical Alzheimer. I call that Alzheimer. Then there's MCI due to Alzheimer--mild cognitive impairment due to Alzheimer disease. They have amyloid in the brain, they have mild symptoms, but they don't have dementia, they can still take care of themselves. And then I also think about Alzheimer disease as mild, moderate and severe Alzheimer disease where the person can no longer take care of themselves, and that's kind of when I was in med school, that's what I learned Alzheimer disease was. Dementia, they had it, that's when the disease started, we now know, Alzheimer pathology begins in the brain decades, 20, 30 [years], probably more even decades before symptoms begin. So a lot to talk about Marc. But when you think about these two big elephants in the room, maybe, you know the FDA has allowed or given the physician I guess free roam or wherewithal to figure this out, to figure out which patients, which stage, do we need a marker, do we not? What are your thoughts about the amyloid biomarker aspect? And what are your thoughts about what truly is the treatment of Alzheimer disease? What are the stages, when to use it?

Marc E. Agronin, MD: You've really described the major shift that we've seen in the past decade, and that is going from making a good guess on whether someone has Alzheimer disease, to now seeing evidence. I refer to these imaging modality as basically virtual brain biopsies because we get a window into is there amyloid or tau in the brain? And on the one hand it's brought us clarity, it's allowed us to study treatments and individuals who actually have what we're trying to target--they have amyloid in the brain or now with newer studies, they have tau in their brain. So, that's been an important change. It's also as you said, allowed us to recognize that this disease doesn't begin when someone walks in our office, it's been going on a long time. On the other hand, it's proved vexing because we're seeing individuals who are minimally are not symptomatic at all and they have amyloid in the brain. And what does that mean? And now tau, what does that mean? We can't predict the course. And you and I both know, as we follow people literally for the rest of their lives, that we've seen many different courses, we really appreciate the heterogeneity of this disorder. So I have individuals who, if you do nothing, and we know they have amyloid, they tend to do pretty well and drift slowly over the years. And we see other people who really progress very quickly, we see the same thing in these clinical trials. And I would add on top of this, what makes it more confusing is I have people come in, who in past years, I would say absolutely this is Alzheimer disease. They fit the clinical picture. They fit the progression, but they have no amyloid in the brain or they have no tau in their brain. What do we diagnose them with? We can say this as a major neurocognitive disorder, but that doesn't bring much clarity. So, as much as this new era of biomarkers has really helped us to understand better what's going on in someone's brain, it's also brought so many more questions and areas that need to be clarified over time. And now we bring aducanumab into this and the question is, unfortunately, at least this point, the label is not helpful, who would qualify for this? And I think this is a dilemma because as clinicians, you and I can talk about how we make a diagnosis. But we both know that a lot of this is going to be the hands of the payers. And they may say, you have to have a spinal tap, you have to have an amyloid scan, but will they pay for it? So there's a lot yet to be revealed in the coming weeks and months before we really have a sense for who will be able to get this treatment, who will have access to it. So that lies ahead.

Richard S. Isaacson, MD: Yeah, so, I mean, so much to talk about. As you mentioned earlier, the EMBARK study. So the EMBARK study is a phase 3b, open-label, single-arm study, safety study also, to look at not just safety but also efficacy, okay, safety and efficacy of aducanumab, and it's going to be looking at people for two years, it's 24 months, to understand what is truly the long-term efficacy and safety. The other part about, you know, this next phase of real-world clinical research in our clinical practices, you and I and hundreds of other physicians are going to be prescribing this drug at some point. And the real-world impact, kind of like the phase 4 outcomes of how aducanumab works, and what the safety is. This is a very unique situation, because we're going to be collecting data and the FDA said that, you know, the approval is contingent on collection of, you know, further data. And, you know, clinician researchers like us and clinicians are going to, at some point have to report these real-world outcomes, study them in a rigorous way. I guess how do you think a physician should talk about this with a patient? You have an FDA approval, but you also have the need to continue to study whether or not it's effective and safe. How do you deal with that?

Marc E. Agronin, MD: This is a big challenge. It's important that the EMBARK study continues, because we need to continue to gather data on individuals on aducanumab and it's open-label. These are individuals who've been on it before. And so we'll see over time, whether and I think the FDA has even signaled that this data is going to be really clear. But you know, it's going to be clear in terms of helping to determine to some extent the fate of aducanumab, but the problem is, once you've let it, you know, the horse out of the barn, can you really put it back? And you know, can you imagine if the data from EMBARK is not as positive as the other data a year or two years from now, and already have hundreds if not thousands of people on the treatment? What will you do? And so I think that's going to be a challenge. You know, the other issue is, how are we going to identify these individuals. You and I know that the diagnostic process, you really have to take it seriously, these are not just shooting from the hip diagnoses. And my concern is that there's going to be individuals who are going to come into their primary care doctors and other individuals who this is not what they do every day, and there's going to be a lot of pressure to just put them on this treatment. Will the diagnostic process be as careful if it's not outlined, now? It's not in the package label? What will it entail? So I think there's ]a lot ahead that we really need to determine. I'm also concerned about what will be the impact of this on other research studies. You know, at our center MIND Institute at Miami Jewish Health, we conduct all the major Alzheimer disease trials. And already I'm hearing from subjects who say, well, you know, why would I go into a placebo-controlled trial if I can get the real thing? At the same time, we know and I know from as an investigator that, in general, with these studies, if you're not rigorously looking at the metrics of it, it can be hard to really tell what's going on with someone. And so if we're doing this treatment in the real world, it's approved, and we're not really collecting any data on it, how do we know we're slowing it down? And so you know it would be nice if somehow we can build into the real world use of this some set of metrics that we can follow over time, and this data can help over time. I think that's really important and I think that will help address some of the concerns that I'm hearing from a lot of clinicians out there on this.

Richard S. Isaacson, MD: I mean, you know, I have a very unique practice as you know, an Alzheimer's Prevention Clinic. We only see people in the pre-dementia phase, so we see preclinical Alzheimer patients that we've identified as having amyloid, but no symptoms or minimal symptoms that are not even close to MCI mild cognitive impairment. We see people with MCI mild cognitive impairment, no dementia, they can still go about their activities of daily living. So we already have this very robustly, well-characterized group of people--very unique, though, not very common. The question that I have is, how do you identify the appropriate patient? How do you in clinical practice, figure out who should go on aducanumab because the label is so broad. The treatment of Alzheimer disease to me, if I'm, you know, looking at the definition of Alzheimer, I include Stage 1, 2, and 3. But the clinical data, sorry, the research data, only really looked at people with MCI and early Alzheimer disease in terms of their Mini-Mental and in terms of their functioning. The other part about how I look at who was the appropriate patient? It's not just about the efficacy part; in which patient group should this drug be used to promote efficacy, which is I guess, cognitive maintenance or cognitive improvements or reduction of amyloid? What is the definition of efficacy? That's complicated. But then who is the appropriate patient from a safety perspective? So in the Alzheimer's Prevention Clinic, we're already doing cognitive tests every six months. Okay, well, that's helpful, because we can understand trajectory. In the Alzheimer's Prevention Clinic, we're doing APOE on everybody, because APOE4 variant increases risk a little bit--not good for diagnosis, necessarily, but we use it to help personalize care. Complex, a little controversial, but we use the APOE4 variant, and it's helpful for us. I know that when a person with the APOE4 variant, actually, when they have two of them, they may be at higher risk for ARIA, for vasogenic edema, for bleeding, maybe we need to be more cautious with the dosing, maybe we need to have a different titration schedule. We have a label that says treatment of Alzheimer disease, we have a label that says, you know, do an MRI at seven months and 12 months. Well, what if a person has 2 E4s versus E3? So we have a label that is so broad. So I guess this is again, a lot to talk about. But what is the appropriate patient from an efficacy and a safety perspective based on the data and based on your feeling?

Marc E. Agronin, MD: Well to start with the efficacy front, you know, the work you've done has really been pioneering in terms of looking at prevention. And this is we know, we need to start early and identify those individuals. And ideally, those are the individuals who we treat with this, because going back to my analogy before about, you know, the fire in the house, you want to put it out before there's a lot of damage, you know, the brain does not repair itself as readily and as quickly as other parts of your body, we just know that. And so, theoretically the earlier we catch it and rid the brain of amyloid and tau, we would imagine, now this is not based on the data we have yet, but we would imagine and want to get emergent data showing that that will actually push off the emergence of this disease, it will really slow it down. But we yet we still have to establish that. So on the one hand, those theoretically are the ideal patients, we need more data to really support that. But you know, years before they get symptoms, that's when we should be using this, but we haven't really identified them or diagnosed them formally as Alzheimer disease. So you've identified the key challenge. And most of the calls I'm getting are from individuals or loved ones of individuals who have moderate to severe disease, because they're willing to do anything, even if this is a Hail Mary and somehow would slow it down. And yet we know that they will not be appropriate because there is no data showing it's going to have an impact at that point. So the efficacy thing, I think will be straightened out over time, I think it'll be a combination of clinicians and, you know, working with the carrier, the you know, the payers, the insurance carriers to determine what's even to be covered in the first place. I am concerned about access to, you know, diverse communities, because I think that there's going to be a greater challenge there. And the other thing is that sites, you know, centers like you, centers like me, you know, we're a not-for-profit site; in terms of being able to give these treatments, sometimes it's more challenging, and there's more barriers. And, you know, the infusion sites in the community that we have to partner with, there's just a lot to be worked out. Safety is also important. On the one hand, I can tell you from being an investigator for every major immunotherapy trial over time that we definitely see ARIA, you know, these amyloid-related imaging abnormalities, whether it's a vasogenic edema, whether it's these micro-hemorrhages, you know, most of the time it's asymptomatic. When it's symptomatic, it's tends to be pretty mild. That's the good news about this. I think, you know, some of the accounts I've been reading, you know, tend to describe it a little more severe than it typically is. But we usually if we catch it, it's usually earlier than seven months in some individuals. So I'm a little, I have a question about waiting that long. And what if someone doesn't want to get an MRI? Or what if someone can't get an MRI for some reason? What about individuals who, you know, because of a pacemaker or you know, some metal in their body? You know, they have early Alzheimer disease, but you can't do an MRI, so are they now ineligible for this, what do you do? So those safety issues also need to be clarified. I think what you're pointing out very clearly here is that, you know, in this first week, amidst a lot of hoopla, and a lot of hope, there's so many questions, and so many issues or concerns being raised that we really have to take our time to address these before rushing into, you know, wholesale treatment of individuals with this new agent.

Richard S. Isaacson, MD: Yeah, I agree. You know, night one, first night, from 7:45 to at least 10:00PM I was getting back to patients. I treat a lot of MCI patients. I have a lot of preclinical Alzheimer patients; the patients I was calling back on that Monday night, every single one invariably moderate to severe disease. Those are the patients that were calling, those were the patients that were emailing, they wanted to know, you know, what can we do, I will rent an apartment in New York because we can travel back and forth. If this is going to be it, we'll get a studio, like we'll do whatever it takes. So you're exactly right, the people that are most interested at this exact moment and in those days and weeks soon after, will be people that haven't really been shown to be effective, where the drug hasn't been tested and shown to be effective.

Marc E. Agronin, MD: That's another concern of mine, because there are people who, you know, for good reason they're desperate. It's been, you know, almost 20 years since any approved treatment, and they're willing to pay money and do what they need to do. And you and I both see they do things sometimes that have no scientific validity and they're exploited at times and they end up spending a lot of money on treatments that there's no evidence for. And I'm concerned in this case, if there'll be areas in which people are just going to pay just to get it done even in those stages where it's not supportive.

Richard S. Isaacson, MD: And the other side of the coin, which I think we'll differ for conversation at this time--maybe we can talk about this offline--but preclinical Alzheimer disease. It's never been studied. We don't really have data you know, we have the A4 study using solanezumab in a preclinical Alzheimer population. We have other studies the Ahead 345 study, which is going to be using lecanumab or BAN-2401 which was the old name, looking at more the preclinical population. Actually they even define it as early preclinical and I think later preclinical something like that, you know? Is this going to be used for people with preclinical Alzheimer disease? When do you pull the trigger? That's off-label right? Or is it off label because it says the treatment of Alzheimer and preclinical Alzheimer disease is invariably Alzheimer disease, or is it?

Marc E. Agronin, MD: That's a key point: what's on-label, what's off-label because we know how the diagnoses it's varied over time. And we make clinical diagnoses without biomarkers all the time. So does that count? There's no biomarker specifications.

Richard S. Isaacson, MD: So it's going to be an interesting several weeks and an interesting several months, I think for sure.